High Dynamic Range Imaging Signal Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Low dynamic range image sensors struggle to achieve high dynamic range imaging without introducing motion and flicker artifacts, making it impossible to decode the integration time signal combination used for constructing pixel HDR values, which affects the reliability of applications like machine vision and ADAS.
Innovation Solution
A method and apparatus that select the 'best' signal for each pixel location based on non-saturation and integration time to construct an HDR output, ensuring accurate color representation by choosing the non-saturated signal, the signal with the greatest value, or the shortest integration time when multiple signals are available.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Illumination intensity
If multiple integration time signals are combined to achieve high dynamic range, then the dynamic range is improved, but motion and flicker artifacts are introduced that compromise color accuracy
Solution Approach 1:
The patent segments the pixel array into multiple independently controllable regions, each capable of using different integration time signals. This allows selective application of multi-integration-time processing only where needed (in high dynamic range regions) while avoiding it in uniform regions, thereby preventing motion and flicker artifacts in areas where they would compromise color accuracy.
Solution Approach 2:
The patent applies different processing strategies to different regions of the image based on local characteristics. High dynamic range regions undergo multi-integration-time signal combination, while uniform regions use single-integration-time processing. This local differentiation ensures that color accuracy is maintained in regions where multi-integration processing would introduce artifacts.
2Illumination intensity
If multiple integration time signals are combined to construct HDR output, then the dynamic range is improved, but the integration time signal combination information is lost making applications unreliable
Solution Approach 1:
The patent performs preliminary encoding of integration time signal combination information into the HDR output data structure before the data leaves the sensor. This preliminary action ensures that the information is preserved and transmitted along with the image data, allowing applications to later decode and use this information for reliable operation.
Solution Approach 2:
The patent introduces an intermediary encoding mechanism that bridges the gap between the multi-integration-time processing and the final HDR output. This intermediary layer embeds the integration time signal combination information into the output data structure, serving as a mediator that preserves critical information that would otherwise be lost in the combination process.
3Illumination intensity
If linearization is applied to normalize signals to longest integration time, then the dynamic range is improved, but motion artifacts are introduced that alter actual color of moving objects
Solution Approach 1:
The patent dynamically selects which integration time signals to combine and how to process them based on local image characteristics and motion detection. This dynamic approach allows the system to adapt to moving objects by adjusting the processing strategy, thereby reducing motion artifacts while maintaining the benefits of extended dynamic range.
Solution Approach 2:
The patent changes processing parameters such as integration time selection and combination weights based on local image content and detected motion. By adjusting these parameters dynamically, the system can minimize motion artifacts in regions with moving objects while still achieving high dynamic range in static or less critical regions.
Data Source
AI summary
Various embodiments of the present technology comprise a method and apparatus for high dynamic range imaging. According to various embodiments, the method and apparatus for high dynamic range imaging is configured to select the “best” signal for each pixel location to construct an HDR output. In one embodiment, the “best” signal is select among various pixel signals and the pixel signal selected as the “best” signal is based on the value of the pixel signals and identification of non-saturated. If only one pixel value for a particular pixel location is non-saturated, then the “best” signal is the non-saturated signal. If more than one pixel value is non-saturated, then the “best” signal is the pixel signal with the greatest value. If two more pixel values are non-saturated and both have equally great values, then the “best” signal is the pixel signal with the shortest integration time.